Application Value Of DCE-MRI Quantitative Parameters For Non-invasive Assessment Of Liver Fibrosis In Children With Congenital Choledochal Cyst

Objective:To investigate the value of dynamic contrast-enhanced magnetic resonance imaging?DCE-MRI?quantitative parameters in diagnosing and staging liver fibrosis caused by congenital choledochal cyst?CCC?in children,to explore a new method for noninvasive diagnosis of liver fibrosis.Materials and Methods:Totally 52 children with CCC were included into case group,while 21 children were selected as control group.All subjects underwent liver DCE-MRI with Siemens MAGNETOM Skyra3.0T,including dynamic contrast-enhanced images and T1 mapping.Among them,dynamic contrast-enhanced images used three-dimensional volumetric interpolated breath-hold examination?3D-VIBE?,7-9s per period,and 40 consecutive scans.At the end of the 4^th period,the group was injected with Gd-DTPA-BMA contrast agent.For T1 mapping,double flip angles T1-3D-VIBE was used.Images were processed using Siemens Syngo.Via post-processing workstation.After fitting into Tofts modes,the following perfusion parameters were obtained:volume transfer constant of the contrast agent(K^trans),reverse reflux rate constant(K_ep)and volume fraction of EES?V_e?.Within one week after the scan,all children in case group were executed choledochal cyst resection and Roux-en-Y hepaticojejunostomy and their livers were kept for HE and MASSON staining.Liver fibrosis staging according to METAVIR staging system,then all children were divided into three subgroups:no?F0?,mild?F1-2?and severe fibrosis?F3-4?.The DCE-MRI data were analyzed by SPSS21.0statistical software package.Normality test and homogeneity test of variance for all data.Univariate analysis of variance was used to compare the quantitative parameters of different liver fibrosis subgroups.The relationship between quantitative parameters and liver fibrosis stages analyzed by Spearman rank correlation analysis.The area under curve?AUC?,sensitivity,specificity,Yoden index and the best cut-off value obtained by receiver operating characteristic curve?ROC?was to evaluate the diagnostic efficacy of each parameter in liver fibrosis.P<0.05 statistically significant.Results:1.The number of cases of each liver fibrosis stage in the case group is as follows:F1,n=12;F2,n=12;F3,n=22;F4,n=6.21 subjects in the control group were classified as F0 based on normal radiological examination and liver function laboratory tests.2.K^trans,K_ep and V_e in the case group were all lower than those in the control group,the difference was statistically different?P<0.05?.3.In the correlation analysis,K^trans and K_ep were negatively correlated with the degree of liver fibrosis?r=-0.729,-0.620,P<0.001?.With the progress of the liver fibrosis?F0-4?,K^trans and K_ep gradually decrease.V_e was also negatively correlated with liver fibrosis stages?r=-0.343,P=0.003?,but weaker than K^trans and K_ep.4.Using One-Way ANOVA analysis,K^trans,K_ep and V_e were significantly different among all fibrosis groups?P=0.000,0.000 and0.023?.Comparing different groups,the difference of K^trans and K_ep in no?F0?vs mild?F1-2?,no?F0?vs severe?F3-4?,mild?F1-2?vs severe?F3-4?fibrosis were statistically significant?P<0.05?.Except mild?F1-2?vs severe?F3-4?fibrosis,The difference of V_e in no?F0?vs mild?F1-2?,no?F0?vs severe?F3-4?liver fibrosis were statistically significant?P=0.015,0.016?.5.Using ROC analysis,the area under curve?AUC?for identifying no vs fibrosis,no vs mild,no vs severe,mild vs severe fibrosis and mild vs advanced fibrosis groups were 0.875,0.788,0.951 and 0.775 respectively by K^trans.The AUC for identifying no vs fibrosis,no vs mild,no vs severe,mild vs severe fibrosis were 0.805,0.698,0.896 and 0.760 respectively by K_ep.And the AUC for identifying no vs fibrosis,no vs mild,no vs severe fibrosis were 0.730,0.675 and 0.777 respectively by V_e.Conclusion:DCE-MRI quantitative parameters K^trans,K_epp and V_e have great value in distinguishing normal liver tissue and liver fibrosis and liver fiber grading,and The diagnostic efficiency of K^transrans and K_ep is obviously better than Ve.DCE-MRI quantitative parameters can be used as one of the reliable assessment indicators for non-invasive diagnosis and staging of liver fibrosis in children.